KR20000042931A - Method for switching duplication according to superiority of errors in board - Google Patents

Abstract

PURPOSE: A method for switching duplication according to superiority of errors in a board, is provided to activate a board which are normal or less wrong, by comparing with a relative board state in generating, by dividing errors to be generated in the board into 3, and then giving the divided errors the superiority. CONSTITUTION: A method for switching duplication according to superiority of errors in a board, comprises the steps of: checking states of a clock, switch, and link of a present board; deciding whether a clock error is generated in the present board; raising a clock error count; deciding whether the a clock error counting value is beyond limits; deciding whether a clock state flag is normal; setting up the clock state flag as abnormal; executing a duplication switch to the present board into other, in duplication; deciding whether a switch or link error is generated in the present board; raising a switch error count or a link error count, if normal; deciding whether the switch or link error counting value is beyond the limits; deciding whether the switch state flag or the link state flag is normal; setting up the switch state flag or the link state flag as abnormal; deciding whether a present clock state flag of the relative board is normal; and executing a duplication switch to the normal clock state flag of the relative board.

Description

Redundancy Transfer Method According to Priority of In-Board Error

The present invention relates to a duplication switching method when an error occurs in an asynchronous transmission mode switch, and more particularly, to a duplication switching method according to a priority of an error that may occur in a board.

In the conventional redundant switching method, as shown in FIG. 1, a redundant switching is performed by diagnosing a state of received data on a chip itself rather than a software switching. Referring to FIG. 1, when the first board 110 continuously generates an error of 5 cells or more, and the second board 120 is in a normal state (S1), the active board is the first board. Transfer from the 110 to the second board 120 is made. For example, in case of Multi-link Transmitter & Receiver Integrated (MTRI), which is Link AI, when one of the receiver outputs is selected and error occurs more than 5 cells in a row, the other party receives an error more than 5 cells. If not, the output of the other receiver is selected. In addition, when one error occurs in the second board 120 and one error occurs in the first board 110, the second board 120 is in the case where the second board 120 is in a normal state (S2). Will continue to operate. However, when the second board 120 has an error of 5 cells or more continuously, and the first board 110 is in a normal state (S3), the operation board is the first board 110 in the second board 120. ) Is made. In addition, when one error occurs in the first board 110 and one error occurs in the second board 120, the first board 110 when the first board 110 is in a normal state (S4). Will continue to operate.

However, since all of the conventional redundant switching methods are simple switching in a chip, there is a limit that intelligent switching cannot be performed by comprehensively checking various errors.

Accordingly, an object of the present invention is to provide a method for performing redundancy switching according to a priority when an error of a duplicated board occurs.

Another object of the present invention is to provide a method for performing redundant switching by comprehensively determining all possible errors in a board.

Still another object of the present invention is to provide a method of switching a board according to a severity of an error currently occurring in a counterpart board.

In order to achieve the above object, the present invention prioritizes errors that may occur in a board so that when a specific error occurs in the current board, the priority of the corresponding error is higher than that of the error occurring in a redundant counterpart board. It is characterized by performing only board duplication transfer.

1 is a state diagram of a conventional redundant transfer.

2A and 2B are diagrams illustrating a redundancy switching process based on a priority of an error occurring in a board according to an embodiment of the present invention.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings. In addition, in the following description of the present invention, if it is determined that a detailed description of a related known function or configuration may unnecessarily obscure the subject matter of the present invention, the detailed description thereof will be omitted.

In the present invention, possible errors in the board are divided into three groups. That is, all errors that can occur in the board according to the present invention are classified into one of a clock error, a switch error, and a link error and processed. The clock error corresponds to a clock sync loss or a clock status loss. There are many causes of switch error. EHEC error is an error in the value of Extended HEC in the cell header, CRC error is also an error related to information corruption in the cell header, and SHCRC error is a common memory. An error occurs in the retrieved cell information, an AFCRC error occurs when an error occurs in a value stored in the AFIFO, and finally an IAPCRC error corresponds to a case where an error occurs in a value stored in the IAP. A switch error is when one or more of these five errors occur. Finally, link errors are cases where synchronization loss, decoding errors, or THECE (transmission data errors) occur during link state monitoring. The priority of these errors is elaborated as follows. The highest priority error is the clock error, which is a serious error condition on the board. Next, switch errors and link errors are given the same level of priority. Therefore, the priority system in the board according to the present invention has a two-layer priority structure. If one of the above three errors is currently occurring on the current board, whether or not to perform a redundancy switchover is compared to the other boards. Allow the board to operate. This principle of redundancy can be achieved by using the software developed in Linkage. In other words, when the software activates a signal called Application fail, the link sets the specific bit in all cell headers as an error and transmits it to the receiver of the other link. Only by accepting it is possible.

In the redundant switching method according to the priority of the error occurring in the board according to the present invention, the following operations should be performed periodically.

First, check for any error conditions that may occur in the board. That is, the current clock state, switch state, and link state are monitored. Then, it is determined whether a clock error having the highest priority among errors is currently occurring. If the clock is normal, the routine proceeds to performing the redundancy switchover depending on the next switch and link failure. However, if the current clock is abnormal, first it is determined whether the clock was normal in the entire period. The reason for this is that redundant switching always runs only when the state transitions from normal to abnormal. If the state of the clock is abnormally changed from normal to abnormal, the application fail signal is driven so that the other link receiving unit performs switching to actively select transmitter data of another redundant board. If the clock was already abnormal before, redundancy switching is not performed and the next priority is to switch and link error handling.

The current switch on the board does not always perform redundancy switching when a link error is detected, even if the previous state was normal. In this case, check the current clock status of the redundant counterpart board and perform redundant switching only when the counterpart's current clock status is normal or when the counterpart's current clock status is abnormal and the current clock of the current board is abnormal. This is because the board with the low priority error becomes the active board. That is, for example, if the clock is normal and the switch or link error is occurring and the counter board becomes an active board due to the redundancy switching despite the abnormal clock on the counter board, this has a greater adverse effect on the quality of service. Results.

2A and 2B illustrate a process of performing redundant transfer by priority of an error occurring in a board according to an embodiment of the present invention.

2A and 2B, a redundant switching method based on a priority of an error occurring in a board according to an embodiment of the present invention will be described. In operation 201, the current state of the clock of the operation board currently in operation is checked. In operation 203, the current state of the switch of the operation board is checked. In operation 205, the current state of the link of the operation board is checked. Check In step 207, it is determined whether a clock error of the operation board has occurred. If it is determined that the clock error of the operation board has not occurred, in step 209, the clock state flag of the operation board is set to normal.

On the contrary, if it is determined that a clock error of the current operation board has occurred, the clock error count of the operation board is increased in operation 211. In operation 213, it is determined whether the clock error counting value of the operation board is greater than or equal to the threshold value. If it is determined that the clock error counting value of the operation board is greater than or equal to the threshold, it is determined in step 215 whether the clock state flag of the operation board is normal. If it is determined that the clock state flag of the operation board is normal, in step 217, the clock state flag of the operation board is set to abnormal, and in step 219, redundant switching is performed.

In operation 221, it is determined whether a current switch or link error has occurred. If it is determined that no current switch and link errors have occurred, in step 223, the switch state flag is set to normal, and the link state flag is also set to normal. In contrast, if it is determined that a current switch or link error has occurred, in step 225, the switch error count or the link error count is increased. In step 227, it is determined whether the switch or link error counting value is greater than or equal to the threshold. If it is determined that the switch or link error counting value is greater than or equal to the threshold, it is determined in step 229 whether the switch state flag or the link state flag is normal. If it is determined that the switch state flag or the link state flag is normal, the switch state flag is set to abnormal or the link state flag is set to abnormal in step 231. In operation 233, it is determined whether the current clock state flag of the opponent board is normal. If it is determined that the current clock state flag of the opponent board is not normal, it is determined whether the current clock state of the current board is normal in step 235. If the current clock state flag of the opponent board is normal or if the current clock state of the current board is not normal, redundancy switching is performed in step 237.

As described above, the present invention classifies three types of errors that can occur in the board in performing the transfer of the redundant boards and gives priority to them again so that the error is always normal or error compared to the state of the counter board. Make the less serious board active. Therefore, the present invention has the effect that the service can be intelligently controlled to increase the reliability of the cell switched data.

Claims (3)

In the redundant switching method according to the priority of the error in the redundant asynchronous transmission mode switch, Prioritize the errors that can occur in the board so that when a specific error occurs on the current board, the board duplication transfer is performed only when the priority of the corresponding error is higher than the priority of the error occurring on the redundant counterpart board. Redundancy transfer method according to the priority of on board error. In a method of performing a redundant transfer according to priority in an asynchronous transfer mode switch, Check the status of the current board clock, switch, and link; Determining whether a clock error has occurred on the current board; If it is determined that the clock error has occurred, increasing a clock error count; Determining whether the clock error counting value is greater than or equal to a threshold; If it is determined that the clock error counting value is greater than or equal to a threshold, determining whether a clock state flag is normal; If it is determined that the clock state flag is normal, setting the clock state flag to be abnormal; Redundancy of the current board to another board, Determining whether a switch or link error has occurred on the current board; If it is determined that a switch or link error has occurred, increasing the switch error count or link error count; Determining whether the switch or link error counting value is greater than or equal to the threshold; If it is determined that the switch or link error counting value is greater than or equal to the threshold, determining whether the switch state flag or the link state flag is normal; If it is determined that the switch state flag or the link state flag is normal, setting the switch state flag or the link state flag to be abnormal; Determining whether the current clock status flag of the other board is normal; And if it is determined that the current clock state flag of the counterpart board is normal, performing a redundancy switching process. The method of claim 2, If it is determined that the current clock state flag of the counterpart board is not normal, redundancy according to the priority of the on-board error, characterized in that further comprising the step of performing a redundant switching if the current clock state of the current board is not normal Transfer method.